A False Panacea

by Nicholas Lemann
POLITICS, MARKETS, AND AMERICA’S SCHOOLS by John E. Chubb and Terry M. Moe. The Brookings Institution, $28.95/$10.95.

THE LEVEL OF liberal intellectual certainty in this country peaked in late 1964 or early 1965. It was still possible at that moment to believe that Keynesian economic-management techniques had permanently solved the problems of recession and unemployment; that the 1964 Civil Rights Act was going to eliminate America’s racial troubles; that anti-communist, interventionist internationalism was the one true path in foreign affairs; and that the federal government was a powerful, fine-tuned instrument for the achievement of liberal goals. In recent memory the government had ended the Depression, beaten Hitler, regulated away the most fraudulent and unsafe excesses of capitalism, built highways across the country, and sent men into space, among other achievements. At the level of everyday life there were all sorts of routine things that our governments did well—deliv - er the mail, maintain public safety, and, most important, operate a system of universal public education that was still the envy of the world.

Then all in a rush came the escalation of the Vietnam War, the Watts riot, and several other profoundly certaintyeroding events. One of them was the release, in July of 1966, of the Coleman Report, a large study, mandated by the Civil Rights Act and funded by the federal government, of the differences between public schools for blacks and for whites. The idea was that the report would show the nation’s schools to be separate and unequal, with black schools operating on far more limited resources and therefore producing lower student achievement. This would provide justification for another landmark piece of legislation, the Elementary and Secondary Education Act of 1965, which had institutionalized the long-deferred liberal dream of federal aid to local public schools. The act was written so as to guarantee heavy subsidies for big-city ghetto schools; the money, presumably, would quickly bring black student achievement up to the national average.
Instead, James S. Coleman, the sociologist who ran the study, found that the differences in per-pupil spending between black and white schools were not as great as most people thought, and that, anyway, the most important determinants of student achievement have to do with family and peer group— whereas spending makes almost no difference at all. The Department of Health, Education, and Welfare, which was just beginning to dispense its new monies to school systems, found the Coleman Report profoundly embarrassing, and arranged for the press conference announcing its findings to be held on the Friday afternoon before the July 4 weekend, with the Secretary of HEW absent, in the hope that few reporters would bother to come.
In the short run, the Coleman Report was used as justification for federal busing orders, which were supposed to help black students by changing their educational peer groups. In the long run, it was crucial in helping to introduce into public discourse the idea that public schools don’t work. All through the post—Second World War period alarms about urban public education had been sounded, but the Coleman Report had the force of social science and a government imprimatur. Also, it carried the implication that “throwing money at problems" never helps, which is one of the essential handful of notions that have shaped our domestic politics since the Great Society.
In recent years Coleman has been arguing that private and parochial schools do a better job of educating poor children than public schools, because they have a richer supply of “social capital.”In his most recent fulllength book. Public and Private High Schools: The Impact of Communities (1987), he and his co-author, Thomas Hoffer, suggested “with some hesitation" that public school students be offered a wider choice of schools to attend on the government’s dime. Nowcome John E. Chubb and Terry M. Moe. with no hesitation at all. proposing that public education is so fundamentally flawed a concept that the system we’ve had for all these years must be replaced by one in which the government funds students, not schools. As they accurately note, “Just a few years ago, the suggestion that choice might succeed in restructuring American education would have been regarded as pure fantasy. Times have changed.”
Chubb and Moe are intellectual descendants of Coleman (and also of the political scientist Charles E. Lindbiom, to whose major work, Politics and Markets, their title pays tribute), but their work, much more than their predecessors’, bears the stamp of the 1980s. Hits of Reagan-era jargon, such as “supply side” and “decontrol,”are scattered through their book. More broadly, although they address only the issue of education, their line of argument assumes that the whole American central government is essentially a spent force—that, as President Reagan said in his first inaugural address, “government is the problem.” In the nation’s intellectual history Politics, Markets, and America’s Schools might one day prove useful in demarcating the moment when the prevailing assumptions about government became precisely the opposite of what they had been a quarter century earlier.
Conservative think tanks like the Heritage Foundation have been promoting educational choice for years; what gives Chubb and Moe’s work its impact is that its publisher is The Brookings Institution, the leading supplier of ideas and personnel to Democratic Presidents. The heavy publicity (for this kind of book) that attended the release of Politics, Markets, and America’s Schools was reminiscent of the reception given the Coleman Report: what made it news was the association with a liberal institution, the dramatic dismissal of the current wave of education reform as being doomed to failure, and the sense that the conclusion reached was a matter not of the authors’ ideological preference but of the inexorable findings of a cast research project. The criticism of the book by mainstream educators has had a sputtering, how-dare-they quality; now Chubb, the front man of the pair, is widely quoted on television and in newspapers as a leading education expert.
As IS OFTEN the case when a book takes on a life of its own in political debate, what’s actually in Politics, Markets, and America’s Schools is somewhat out of sync with its reputation. The first third of the book is a philosophical tract explaining with airtight libertarian logic why public schools can’t possibly succeed and private schools can’t possibly fail. The problem inherent in public schools, according to Chubb and Moe, is not that their standards aren’t high enough but that they operate under “direct democratic control” (a catch phrase the authors use often, and always disapprovingly). Inevitably, direct democratic control generates bureaucracy, and bureaucracy eats away at the autonomy of individual schools by imposing on them “higher-order values”—requirements of one kind or another. (The educational reforms of the 1980s won’t work, Chubb and Moe say, because they merely introduce more rules—for example, teacher-competency standards.) Since autonomy is the one essential precondition of a good school, direct democratic control and quality education simply can’t coexist. Private schools, on the other hand, are controlled by the market, in the sense that each one has to attract a voluntary student body; not coincidentally, they are autonomous, and good.
In the next section of the book Chubb and Moe undertake to provide statistical proof that their theory is correct. Here they enter the realm of quantitative social science, with its heavy emphasis on regression analysis and its impenetrable prose style. Most lay readers will be likely to skip a few chapters ahead.
A concomitant temptation is to accept on its face Chubb and Moe’s assertion that, as they wrote recently in The New York Times, “a nationwide study we recently conducted of more than 400 schools and some 20,000 students, teachers, and principals suggests that poor performance is not really a school problem. It is a system problem.” The temptation should be avoided; the evidence isn’t overwhelming.
Chubb and Moe’s nationwide study consisted not of visits to schools but of running the computer tapes from a “data set” called High School and Beyond, which also provided the evidence for James Coleman’s most recent book. The High School and Beyond data come from the results of an hourlong standardized test administered twice (in 1980 and in 1982) to a large group of high school students, and from the answers to a questionnaire that was distributed to principals and teachers in 1984. When Chubb and Moe are discussing the recent vogue for requiring more student-achievement testing (which to them is yet another restriction on school autonomy), they dismiss standardized-test results as unhelpful to policy-makers.
Statewide testing of students can . . . be a misleading and counterproductive means of evaluating the performance of schools. . . . The reality is that people look at a school’s average test scores and jump to conclusions—unwarranted conclusions— about school performance, with farreaching consequences for their assessments of problems and solutions.
Nonetheless, they have relied completely on test scores for evidence to support their own sweeping conclusions. Also, the impression that emerges from reading through their repeated reformulations of the data is that it was actually something of a struggle to arrive at the inescapable conclusion that public education is fatally flawed. They point out, “The organizational similarities across America’s schools are quite impressive.”
Like most educational researchers from Coleman onward, Chubb and Moe have had to adjust their reform ideas to the fundamental empirical truth that the differences among schools are much less dramatic than the differences among students. Their regression analyses consistently show that how well students do in school is determined to a great extent by factors having nothing to do with school — achievement correlates highly with family income, education, values, and socioeconomic status, and with the student’s ability. When public schools are located in high-income, high-status areas—in shorthand, the suburbs— the structural problems inherent in direct democratic control seem magically to melt away. The one school policy that correlates most highly with student achievement, academic tracking, is a tool available only to principals who have a substantial number of highly motivated, well-prepared students. To put it another way, only inside the city limits, and only in public schools whose students come from poor families that don’t stress education and don’t impose much discipline, can public education fairly be characterized as not working. Private schools almost never draw upon a poor urban constituency, so it is hard to demonstrate that they work simply by virtue of being free from direct democratic control. Bad public schools and difficult-to-teach students appear to have a relationship quite like that between the chicken and the egg.
The explanation that Chubb and Moe give for the success of suburban public schools is that “schools with homogenous, problem-free environments” can work under direct democratic control, because only when educational circumstances are difficult do bureaucracies feel the need to entrench for protective purposes, and to impose bothersome rules. Presumably the homogenous-environment theory would also explain the success of schools that have no autonomy at all. like the ones in the pre-Vatican II American Catholic system, or in present-day Japan, where the Ministry of Education controls and coordinates all the weekly lesson plans. If Chubb and Moe’s basic theory applies only under certain circumstances, that waters it down quite a bit; however, the authors argue that since an affluent student body isn’t an option for most inner-city schools, switching from democratic to market control is the only means available by which to engender real improvement.
Many of Chubb and Moe’s findings about what makes schools good sound right. They say that effective schools tend to stress academics and to set high standards; to have strong principals who aren’t looking for promotions to headquarters jobs at the Board of Education. and teachers who are not just credentialed, tenured time-servers; and (interestingly, in light of the findings of the Coleman Report) to have above-average financial resources. All this would seem to fit right into a body of valuable recent writing about how to make bureaucracies more effective— for example, the work of the defensereform movement, and of corporatemanagement gurus like Thomas Peters and William Ouchi, who consistently find in the market system most of the organizational problems that Chubb and Moe lay at the feet of direct democratic control. The assumption of all these writers is that big bureaucracies are an essential part of modern society, so we have to figure out how to make them work better; nobody ever calls for completely abolishing the Defense Department or big corporations. It’s quite a leap for Chubb and Moe to say that their well-founded complaints about the big-city school systems constitute an unassailable case for the obsolescence of public-educational bureaucracies.
THE LAST SECTION of Politics, Markets, and America’s Schools is the place where Chubb and Moe lay out their own education-reform plan. It is essentially a voucher system, though they insist it’s not, because public funds are never paid directly into parents’ hands—the government would create a tuition stipend for every student, which would be given to the school the student chose to attend. Students with “special needs” (that is, from poor families) would get larger stipends than everybody else, so that schools would have an incentive to take them in. But any school could reject any applicant. Private schools could easily qualify to receive the government funds, and no public school would be guaranteed a student body. The credentialing of teachers would be reduced to a minimum, and the strength of the teachers’ unions reduced to the point that principals could hire and fire at will.
There’s something facile about totally dismissing all education-reform schemes that Nave ever really been tried, while presenting as flawless one that exists only in the theoretical realm. It’s doubtful that Chubb and Moe’s choice plan, which as they present it has a nice fresh sheen (“Without being too literal about it, we think reformers would do well to entertain the notion that choice is a panacea,” they write), could withstand the withering intellectual scrutiny to which they subject the present public education system.
One clause of Chubb and Moe’s plan is a ban on parental donations to schools that receive the government tuition stipends; like their denial that the plan is a voucher system, this is one of several aspects of it that seem designed to allay the obvious liberal objections to it. Since all the country’s elite private schools, and many of the best Catholic schools, operate at a loss and make up the difference through elaborate fund-raising efforts, presumably the entire top tier of the private education system would opt out of the Chubb-Moe plan rather than give up the right to put the touch on parents and alumni. The students who are in those schools now would stay in them, and their government stipends would go unspent. Students who are now in suburban public schools would presumably stay there, with their educational lives substantially unaffected by the switch to a choice system. The question is what would happen to the people who are now students in bad urban public schools. Conservative intellectuals have learned to make the case for education vouchers solely in behalf of the ghetto poor (rather than the tuition-burdened lower middle class), so the official test of a voucher system is how much it improves education in the slums.
Chubb and Moe never provide a precise scenario for the restoration of education in the ghettos, but they clearly imply that many bad public schools would simply fold, once the people in their neighborhoods were given the option to go somewhere else. In their place new, better schools would be founded by entrepreneurs, who would be able to survive economically because of the government stipends. (Remember that the best existing private schools couldn’t accept the stipends without agreeing to forsake private donations; that’s why the birth of new schools is an essential part of the plan.) However, it’s possible to imagine less rosy outcomes in the cities. Most students might choose simply to stay in their bad schools, in which case the benefits of choice would be theoretical (more freedom, less coercion), not educational. Another possibility—a likelihood, in my view—is that a lot of diploma mills would spring up in the ghettos to cash in on the new availability of government tuition stipends. Released from all but the most rudimentary certification and accreditation requirements, these schools would be free to be bad— possibly even worse than the schools they supplemented or replaced. The clinics that have come into being as a result of the Medicaid program (another voucher system without actual vouchers), and the trade schools that live on government tuition and job-training grants, do not make very encouraging precedents, and they are much more heavily regulated than Chubb and Moe’s new schools would be. ‘The ready answer of choice advocates to worries of this kind is to accuse the worriers, as Chubb did recently in an interview with a New York Times reporter, of “really saying that poor people were ‘too stupid’ to pick their own schools.” Politics, Markets, and America’s Schools is full of a certain kind of leftwing boiler plate from the 1960s that has recently been adopted by conservatives, in which “empowerment” and “community control” are the obvious solution to the problems of the city slums. Chubb’s remark to the Times was rhetorically a clean put-away shot, but it seems a little over-facile considering that his book contains page after page of evidence that family background is the single most important predictor of student achievement— more important than the quality of the school. If the greatest handicap suffered by low-achieving students is their parents’ impoverishment, poor education, lax discipline, and scant interest in education (which is what Chubb and Moe say), is it really so ridiculous to worry that these same parents might fail to become tough, savvy, demanding education consumers the instant they obtain the right to decide which school gets their children’s tuition money?
The record of community-control plans in public schools—plans that by Chubb and Moe’s lights should help somewhat, because they increase local autonomy—is far from overwhelmingly positive. New York City is now retreating from the community-control policies it adopted in the late 1960s, under which student achievement failed to rise. Chubb and Moe themselves say, “Schools do not seem to benefit in a large or systematic way from direct parent participation.” Choice plans are much more widely in use, and much more clearly successful. Minnesota has a tuition tax deduction and a choice plan within the public school system. Oregon is considering a tuition tax credit. Most big-city school districts have started specialized “magnet schools” to which students can apply regardless of where they live. The main motivation behind these is to use choice as a means of voluntary desegregation (the magnet schools usually Lise racial quotas in picking their students), rather than as a way of eliminating direct democratic control; the main reason they work is that they attract the better students. What has been the fundamental question since the days of the Coleman Report—How can we substantially improve the urban schools that serve the poorest of the poor?—still doesn’t have a once-andfor-all answer.
New York’s District 4, in East Harlem, has the best-known ghetto choice program, and Chubb and Moe sing its praises at some length (while never mentioning the more ambitious but less successful federally financed choice program at the Alum Rock Unified School District, in California). They do say, though, in somewhat Ayn Randish tones, that East Harlem’s success depends on “a small group of visionaries . . . and their hold on power.” In general, the ghetto schools that are producing good results always seem to have a visionary involved—like James Comer in New Haven, or Marva Collins and George Clements in Chicago, or Joe Clark in Paterson, New Jersey— and to be committed to helping students and their parents assimilate into the middle class.
Governments can’t legislate visionaries into existence, of course, but the incentives of the marketplace aren’t guaranteed to produce them either. The best way to encourage their emergence would be to strengthen the hand of school principals as much as possible, especially with regard to hiring and firing teachers, admitting and expelling students, and making the rules that apply in their schools—in other words, giving them what Chubb and Moe call “autonomy.” Chubb and Moe assume that there is an iron link between school autonomy and student choice; I’m unconvinced. There are plenty of examples in this country and elsewhere of government employees—Army company commanders, county agricultural agents, CIA station chiefs—who operate autonomously, and effectively, as employees of big public bureaucracies, because the bureaucracies have been designed around that goal. Why can’t it be true of principals, too?
Caution ought not to be uppermost in our minds in dealing with urban public schools, since on the whole there is little danger of making them worse than they already are. There is something offensive, too, about the idea of an ambitious inner-city parent being essentially forced to send her children to a school where they will get a poor education. But in many big cities this situation doesn’t really obtain any longer, because parents who make the effort can easily arrange to move their children from a bad neighborhood school to a public magnet school or to a parochial school with very low tuition. The presence of educational options of this kind for innercity parents is all to the good; it’s the leap from there to the next step—doing away entirely with the attempt to find some way to provide reliably good, government-operated education to the urban poor—that is perilous.
The prospect of dismantling the institutions of the public sector in an area as important as education does deserve to be treated with caution. Critics of government like Chubb and Moe argue that the instrumentalities of government, once created, tend to become, in the authors’ words, “part of the status quo that the system inherently works to protect,” and so are difficult to change or eliminate after the need for them disappears. We have learned in the past decade that the opposite is also true: in the present American political culture, at least, once a public function is eliminated, it is nearly impossible to get it started again, because taxpayers don’t want to pay for anything that doesn’t benefit them directly.
Let us imagine that Chubb and Moe’s system was enacted, and that it turned out to be like the public school system it replaced in the sense that it worked pretty well everywhere but the inner cities. Let’s say that there the public schools substantially withered away, and were replaced by new private schools supported by government tuition-stipend monies—but that these new schools mostly turned out to be terrible. Let’s say a consensus emerged that the real answer was to rebuild urban public schools as institutions that would help bring poor people into the mainstream of American life. At that point it would be incredibly difficult to persuade the middle class to support a substantial new program that would help only people who don’t pay taxes. Before taking such an immense risk, we ought to make sure we’ve tried every possible means of making bad schools better which doesn’t involve cutting them loose from the webbing of public life. □